INDEX

Explanations

concepts related to memory and documentation processes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.10

1:0.03

2:0.11

3:0.22

4:0.03

5:0.16

6:0.02

7:0.10

8:0.03

9:0.02

10:0.08

11:0.04

Negative Logits

 Volunte

-2.10

Virgin

-1.94

ilst

-1.91

operated

-1.90

Ranked

-1.88

Bec

-1.85

UK

-1.83

Rated

-1.80

competitive

-1.78

obil

-1.75

POSITIVE LOGITS

 acron

2.85

 diagrams

2.79

 answers

2.66

 phrases

2.64

 guesses

2.60

 jargon

2.54

 meanings

2.53

 snippets

2.51

 formulas

2.48

 gist

2.45

Activations Density 0.663%

concepts related to memory and documentation processes

No Comments

No Known Activations